Re: sorting pathnames by basename
mike ledoux <[EMAIL PROTECTED]> writes: > I see this sort of claim fairly frequently from the Perl crowd, usually > accompanied by 'everything you need to know is in perldoc'. Yet every > time I need to figure out what one of these obscure no-letter variables > does, perldoc falls down: > > mwl@voyager:~>perldoc '$|' > No documentation found for "$|". perldoc gives you documentations for perl modules. "$|" is not a module. These variables are documented in the man pages: jabr@vishnu:~ $ man perl PERL(1) Perl Programmers Reference Guide PERL(1) NAME perl - Practical Extraction and Report Language For ease of access, the Perl manual has been split up into several sections: perlPerl overview (this section) perlfaq Perl frequently asked questions perlvar Perl predefined variables jabr@vishnu:~ $ man perlvar PERLVAR(1) Perl Programmers Reference Guide PERLVAR(1) NAME perlvar - Perl predefined variables DESCRIPTION Predefined Names The following names have special meaning to Perl. Most punctuation names have reasonable mnemonics, or analogs in the shells. Nevertheless, if you wish to use long vari- able names, you need only say use English; at the top of your program. This will alias all the short names to the long names in the current package. Some even autoflush HANDLE EXPR $OUTPUT_AUTOFLUSH $| If set to nonzero, forces a flush right away and after every write or print on the currently selected output channel. Default is 0 (regardless of whether the channel is really buffered by the system or not; "$|" tells you only whether you've asked Perl explicitly to flush after each write). STDOUT will typically be line buffered if output is to the terminal and block buffered otherwise. Setting this variable is useful primarily when you are outputting to a pipe or socket, such as when you are running a Perl program under rsh and want to see the output as it's happening. This has no effect on input buffering. See the getc entry in the perlfunc manpage for that. (Mnemonic: when you want your pipes to be piping hot.) -- John Abreau / Executive Director, Boston Linux & Unix ICQ 28611923 / AIM abreauj / JABBER [EMAIL PROTECTED] / YAHOO abreauj Email [EMAIL PROTECTED] / WWW http://www.abreau.net / PGP-Key-ID 0xD5C7B5D9 PGP-Key-Fingerprint 72 FB 39 4F 3C 3B D6 5B E0 C8 5A 6E F1 2C BE 99 Some people say, "The enemy of my enemy is my friend." I often respond, "When elephants fight, it's the grass that gets trampled." msg00420/pgp0.pgp Description: PGP signature
Re: UNIX Arcana [was Re: Perl (or Unix vs. MS, actually) ]
[EMAIL PROTECTED] asked: >>> was Unix ever developed on any of those? (meaning the 12-bit PDP-8/PDP-12 architectures) AFAIK, no. I believe that the original development was on some PDP-11's (11/45's?) that Bell Labs had at the time. Those, of course, are 16-bit machines. But, I don't ever hearing about any "back-porting" to the PDP-8's. At the time, it would have been a real PITA, because the I/O architectures were totally different. Also, by the time that UNIX was more or less working well enough to distribute to its users, the PDP-8 was rapidly approaching end-of-life; DEC had jumped through some really wierd hoops to get it to address 128kb of memory, and it was clear that that was it. The PDP-11s, on the other hand, could do that much fairly easily and was obviously much more extendible. Much of the development in the mid-70's was to meet some interesting requirements that customers had voiced, and obviously some of those customers were running UNIX. HTH, Bayard, who tried, but failed to find his old copy of the RIM loader... ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: sorting pathnames by basename
On Wednesday, August 21, 2002, at 07:52 PM, mike ledoux wrote: > So, if I'm in a rush (usually), and I need to figure out what someone > else's unreadable Perl does, where should I be looking? I'm too lazy to lift my hands off my keyboard or mouse, to the shelf right in front of my face where "Programming Perl" rests, and look up the variables. I usually just head over to http://perldoc.com/ and read the man pages there. For some reason it's just more legible when presented as a web page than as a man page in a terminal window. (Must be the whitespace.) I've bookmarked http://perldoc.com/perl5.6.1/pod/perl.html for easy access. Erik -- Erik Price email: [EMAIL PROTECTED] jabber: [EMAIL PROTECTED] ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: sorting pathnames by basename
On Wed, 21 Aug 2002, at 7:52pm, mike ledoux wrote: > So, if I'm in a rush (usually), and I need to figure out what someone > else's unreadable Perl does, where should I be looking? For variables, try 'man perlvar' (or 'perldoc perlvar', if you prefer). Another very useful reference is 'man perlre' (for regular expressions). There are a whole bunch more. 'man perl' will give you a nice list of topics, and 'man perltoc' will give you a grep'able keyword index. (Yes, Derek, this is not immediately obvious. ;-) HTH! -- Ben Scott <[EMAIL PROTECTED]> | The opinions expressed in this message are those of the author and do not | | necessarily represent the views or policy of any other person, entity or | | organization. All information is provided without warranty of any kind. | ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: UNIX Arcana [was Re: Perl (or Unix vs. MS, actually) ]
On Wed, 21 Aug 2002, at 6:44pm, Jon Hall wrote: > To throw a bit (pun un-intentional) more into this discussion, don't > assume that a "byte" was eight bits. The PDP-8, Linc-8 and PDP-12 for > instance, were all twelve bit words, broken down into two six-bit > characters. I'm just curious... was Unix ever developed on any of those? I was pretty much under the impression that Unix assumes an 8-bit byte, but I don't really have anything to back that up... -- Ben Scott <[EMAIL PROTECTED]> | The opinions expressed in this message are those of the author and do not | | necessarily represent the views or policy of any other person, entity or | | organization. All information is provided without warranty of any kind. | ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: UNIX Arcana [was Re: Perl (or Unix vs. MS, actually) ]
To throw a bit (pun un-intentional) more into this discussion, don't assume that a "byte" was eight bits. The PDP-8, Linc-8 and PDP-12 for instance, were all twelve bit words, broken down into two six-bit characters. Nevertheless, back in those days saving a few bits for every entry in a symbol table was worth the time and effort. md -- = Jon "maddog" Hall Executive Director Linux International(SM) email: [EMAIL PROTECTED] 80 Amherst St. Voice: +1.603.672.4557 Amherst, N.H. 03031-3032 U.S.A. WWW: http://www.li.org Board Member: Uniforum Association, USENIX Association (R)Linux is a registered trademark of Linus Torvalds in several countries. (SM)Linux International is a service mark of Linux International, Inc. ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: Redhat 7.2 install on Alpha
On Wed, 21 Aug 2002, Derek Doucette wrote: > I'm pretty sure that there are a few Alpha people on this list. I'm > trying to install 7.2 on a digital alphaserver 800 5/500. When trying > to boot I get the following: > > IDE Command not successfully terminated > ST0 = 40 > Missing address mark This is the kernel reporting this or the firmware? I've never seen that error before. Is this booted from SRM or AlphaBIOS/MILO? > I have searched google a bit and found some posts with similar problems > but no results. I'm pretty familiar with linux, but this is my first > shot at an alpha box. Any ideas to get me started? Well, I've dealt with Alpha for about 5 years but never seen that one before :) --rdp -- Rich Payne http://talisman.mv.com ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Redhat 7.2 install on Alpha
I'm pretty sure that there are a few Alpha people on this list. I'm trying to install 7.2 on a digital alphaserver 800 5/500. When trying to boot I get the following: IDE Command not successfully terminated ST0 = 40 Missing address mark I have searched google a bit and found some posts with similar problems but no results. I'm pretty familiar with linux, but this is my first shot at an alpha box. Any ideas to get me started? Derek -- NeoMail - Webmail that doesn't suck... as much. http://neomail.sourceforge.net ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: MELBA Tonight
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wed, 21 Aug 2002, Jeffry Smith wrote: > What's the word on a MELBA meeting tonight? I can finally make one, but I > haven't seen an announcement. You must have missed this: On Mon, 19 Aug 2002 [EMAIL PROTECTED] wrote: > When: Wednesday, 21 August 2002, 19:30ish > Where:Martha's Exchange, Nashua (2nd floor) > > Dinner will be at 18:00ish in the usual corner. Walk in the front So, despite the fact that this is the *THIRD* Wednesday of this month, we are indeed meeting tonight. - -- -Matt The goal of science is to build a better mousetrap; the goal of nature is to build a better mouse. -BEGIN PGP SIGNATURE- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE9Y/voc8/WFSz+GKMRAkaiAJ9p5yhYPToVugVBI6eHTAS0fHUdVACgupPT ivYsaynLeI1qKBScoXhDvlg= =Opap -END PGP SIGNATURE- ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: [gnhlug-announce] MELBA Meeting Wednesday night
The usual place would be Martha's Exchange on Main St. in Nashua. Here is a link to the directions: http://wiki.gnhlug.org/twiki2/bin/view/Www/Marthas_directions Dinner is at 6pm. Go in the front door and turn right. Turn left at the wall and go past the micro brewers. The usual table(s) are in the back. Regards, Jeff K. At 04:24 PM 8/21/2002 -0400, Ganesan M wrote: > > > > When: Wednesday, 21 August 2002, 19:30ish > > Where: Martha's Exchange, Nashua (2nd floor) > > Why: Because we haven't gotten together there in a while. > > (and to heckle Ben :) >^^ >Where the heck is this Heckle Ben? Just kidding. >I don't know the usual place. I 'd appreciate >if you are little more specific. > > > Topic: We have none, but if people want to attempt a mini-install > > fest, bring CDs of the distro of your choice, and we'll aim > > to help you install. > > >I 'd like to install mandrake on my laptop. >Does anyone owns Mandrake 8.1/8.2? >I 'd appreciate if you bring it to the meeting >if you have one. > >I already have 8.0. > >G. > > > >___ >gnhlug-discuss mailing list >[EMAIL PROTECTED] >http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: [gnhlug-announce] MELBA Meeting Wednesday night
On Wed, 21 Aug 2002, at 4:24pm, Ganesan M wrote: >> (and to heckle Ben :) > ^^ > Where the heck is this Heckle Ben? Just kidding. Alas, I won't be able to make it tonight, so you will have to find someone else to heckle instead. ;-) -- Ben Scott <[EMAIL PROTECTED]> | The opinions expressed in this message are those of the author and do not | | necessarily represent the views or policy of any other person, entity or | | organization. All information is provided without warranty of any kind. | ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: [gnhlug-announce] MELBA Meeting Wednesday night
> > When: Wednesday, 21 August 2002, 19:30ish > Where: Martha's Exchange, Nashua (2nd floor) > Why: Because we haven't gotten together there in a while. > (and to heckle Ben :) ^^ Where the heck is this Heckle Ben? Just kidding. I don't know the usual place. I 'd appreciate if you are little more specific. > Topic: We have none, but if people want to attempt a mini-install > fest, bring CDs of the distro of your choice, and we'll aim > to help you install. > I 'd like to install mandrake on my laptop. Does anyone owns Mandrake 8.1/8.2? I 'd appreciate if you bring it to the meeting if you have one. I already have 8.0. G. ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
MELBA?
What's the word on a MELBA meeting tonight? I can finally make one, but I haven't seen an announcement. If there isn't one, is anyone planning on going to Martha's anyway for dinner/talk? jeff (who does, in fact, live) ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: UNIX Arcana [was Re: Perl (or Unix vs. MS, actually) ]
On Wed, 21 Aug 2002, at 2:52pm, Bill Freeman wrote: > The description is close. Radix 50 actually allows you to get three > characters into a 16 bit word (40*40*40 <= 65536), or 6 into a 32 bit > word. Ya know, I thought a gain of only one character (five characters, vs the four 8-bit bytes in a 32-bit word) seemed too small, but I didn't actually do the math. Hypothesis: Since "creat" is a system call, and system calls are often implemented "behind the scenes" by functions with an underscore prefix (i.e., "_creat"), that might explain where the sixth character went. This debate only goes to show why Ken Thompson might well believe that the biggest mistake he made was spelling "creat" without the trailing "e". :-) -- Ben Scott <[EMAIL PROTECTED]> | The opinions expressed in this message are those of the author and do not | | necessarily represent the views or policy of any other person, entity or | | organization. All information is provided without warranty of any kind. | ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: UNIX Arcana [was Re: Perl (or Unix vs. MS, actually) ]
Mark Komarinski writes: > Good thing more colors other than green and amber were invented too. Newcommer! We only had black print on those cards and listings. ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: UNIX Arcana [was Re: Perl (or Unix vs. MS, actually) ]
[EMAIL PROTECTED] writes: > Way back when 16 kilobytes was a lot of memory, a method for encoding five > characters into a single 32-bit machine word was developed. It was called > "Radix-50", or "RAD50". The 50 is octal, or 40 decimal. The character set > was 26 monocase letters, 10 digits, three special characters, and a null (a > total of 40). This encoding was used in the linkers of various DEC PDP > operating systems, which is where Unix was born. > > (That could, of course, be incorrect, but I did find references to > Radix-50/RAD50 in some old DEC migration documentation.) It was also known as "squoze code" around the M.I.T. AI lab. RAD50 may have originated at DEC, but I wouldn't bet either way. The description is close. Radix 50 actually allows you to get three characters into a 16 bit word (40*40*40 <= 65536), or 6 into a 32 bit word. 5 characters will actually fit into 27 bits (10240 being no greater than 134217728). That means that you can also have up to 5 bits of "flags", say, for symbol type, in the same (double 16 bit) word of your symbol table as that which stores the name. When it wasn't uncommon to have as little as 4k words (8k bytes) on your PDP-11/20, people really did care about bit twiddling. People writing in assembly language for such machines didn't seem to chaff at such symbol length restrictions. (Perhaps few of them could type, and short names were easier.) Linkers developed for use with assemblers were pressed into service for linking C because originally the C compiler produced assembly languagy, and the assembler was invoked on the output. Bill ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: sorting pathnames by basename
On Mon, Aug 19, 2002 at 05:31:14PM -0400, Kevin D. Clark <[EMAIL PROTECTED]>
wrote:
> You have a list of stuff that you want sorted. The problem is is that
> you want your stuff sorted according to a field contained in the input
> (the last field). Further complicating matters is the fact that this
> field is located at a non-constant place in the input field.
I was thinking that `sort -t / -k -1` would be really cool here, but
sort can't sense of a negative column index. I wonder is the FSF would
accept such a patch to GNU sort?
> The solution: take the input, generate a list from each line, each
> element in the list consists of a tuple consisting of the original
> line plus the comparison key. Sort the tuple-list using the
> comparison key, and then after the sort is done, strip off all of the
> comparison keys, returning the original list ({sans} tuples), sorted.
>
> This is actually a well-known technique in Perl, called the
> Schwartzian Transform. Look it up on the web -- there are plenty of
> good descriptions of it.
That is basically what my example did, but continuing to use sed and
sort instead of perl (for no particular reason -- I could come up with
rationales but they may not be applicable). My suggestion took the
filename, prepended it so sort could get at it, sorted the data by the
first column only, and then removed that column. Since I don't see my
post for some reason, I'll repeat it here:
sed -e 's;\(\(.*\)/\)*\(.*\);\3/\1\3;' | sort -fdt/k1 | sed -e 's;[^/]*/\(.*\);\1;'
Since this is mildly different, and I've never heard of the Schwartzian
Transform, I'm going to propose calling this Bell's Method. :-)
--
Bob Bell <[EMAIL PROTECTED]>
-
"The greatest wage gains between 1997 and last year (1998) were
reported by people working with Linux, a free operating system.
Their knowledge of this increasingly popular system puts them in
high demand for Internet projects..."
-- http://detnews.com/1999/technology/9901/08/01080223.htm
___
gnhlug-discuss mailing list
[EMAIL PROTECTED]
http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: [Tutor] little something in the way of file parsing
On Tuesday, August 20, 2002, at 05:37 PM, [EMAIL PROTECTED] wrote: > Heh. This is pretty ironic. One of the standard Luddite responses to > Python is that whitespace is syntactically significant. Personally, I > find > it a bit weird, but I'm not used to it, and it certainly isn't a > show-stopper. But in this case, this property of Python has come into > conflict with something: The message was mangled by line-wrapping, and > thus, > the indentation is no longer sane. Any chance you can either (1) fixup > the > code to wrap < 75 columns, or (2) post it on the web somewhere? Good call. I talked to the original author (Sean Perry), and he said the code is public domain. I copied it from the origninal email to a text file, it's located at http://homepage.mac.com/~erikprice/debian_avail.py I encourage anyone with the "available" file on their [Debian] system to try it out. I hope that I didn't mess up the whitespace, if I did just let me know. Sean also told me that the script was *not* used in any final Debian release, so if you look for it there you won't find it. But he has been working on improving the script when he's had some free time, so who knows for the future. Erik -- Erik Price email: [EMAIL PROTECTED] jabber: [EMAIL PROTECTED] ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: UNIX Arcana [was Re: Perl (or Unix vs. MS, actually) ]
On Wed, 21 Aug 2002, at 10:10am, [EMAIL PROTECTED] wrote: > Okay, I'll buy that, but why create a linker that only supports 5 > character function names? Okay, some Google searches eventually tracked down this explanation: Way back when 16 kilobytes was a lot of memory, a method for encoding five characters into a single 32-bit machine word was developed. It was called "Radix-50", or "RAD50". The 50 is octal, or 40 decimal. The character set was 26 monocase letters, 10 digits, three special characters, and a null (a total of 40). This encoding was used in the linkers of various DEC PDP operating systems, which is where Unix was born. (That could, of course, be incorrect, but I did find references to Radix-50/RAD50 in some old DEC migration documentation.) > I guess I don't fully understand the role of linkers, if this should be an > obvious thing :) A compiler/assembler turns source code into object code modules. Those modules are not working programs yet; they have symbolic names instead of addresses for variables and functions in many places. You then need to link the modules together, which includes replacing symbolic names with machine addresses everywhere. The output of the linking process is an executable program. Run-time dynamic linking makes things more complex. *hand wave* -- Ben Scott <[EMAIL PROTECTED]> | The opinions expressed in this message are those of the author and do not | | necessarily represent the views or policy of any other person, entity or | | organization. All information is provided without warranty of any kind. | ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: UNIX Arcana [was Re: Perl (or Unix vs. MS, actually) ]
Good thing more colors other than green and amber were invented too. -Mark On Wed, Aug 21, 2002 at 11:00:10AM -0400, Andrew W. Gaunt wrote: > > Back in the early days of computers there weren't > as many characters to go around and folks had to > be very conservative with their use. Since then, more > have been pulled out of the ground so we can use > them more liberally. > > -- > __ > | 0|___||. Andrew Gaunt *nix Sys. Admin., etc. > _| _| : : } [EMAIL PROTECTED] - http://www-cde.mv.lucent.com/~quantum > -(O)-==-o\ [EMAIL PROTECTED] - http://www.gaunt.org ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: sorting pathnames by basename
Bob Bell <[EMAIL PROTECTED]> writes: > Fix your box :-) Point taken, although I try to write portable code, and as such, I will always be a little bit conservative when I am using sed. > > 2: Those options to sort don't work on my Linux box. > > Really? They work on mine. What `sort` do you have? Mine's from > a textutils-2.0e-8 RPM. All options I used are part of UNIX 98 > (http://www.opengroup.org/onlinepubs/007908799/xcu/sort.html). $ rpm -q -f /bin/sort textutils-2.0.21-1 > > Still, I see where you're going here. How about this? > > > > sed -e 's;\(.*\)/\([^/]*\)$;\2/\1/\2;' | sort -t / +0 | sed 's;[^/]*/;;' > > > > This works on the various Unix boxen that I have at my disposal. > > Nope, you've just forced the pathname to contain a '/'. But here's > a simplified version that I think will work around your issues for > non-UNIX 98 compliant machines: > > sed -e 's;\(.*/\)*\(.*\);\2/\1\2;' | sort -f -d -t / +0 | sed -e 's;[^/]*/;;' My example, derived from yours, seems to work, as well as your new solution. Regards, --kevin -- Kevin D. Clark / Cetacean Networks / Portsmouth, N.H. (USA) cetaceannetworks.com!kclark (GnuPG ID: B280F24E) alumni.unh.edu!kdc ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: UNIX Arcana [was Re: Perl (or Unix vs. MS, actually) ]
Back in the early days of computers there weren't as many characters to go around and folks had to be very conservative with their use. Since then, more have been pulled out of the ground so we can use them more liberally. -- __ | 0|___||. Andrew Gaunt *nix Sys. Admin., etc. _| _| : : } [EMAIL PROTECTED] - http://www-cde.mv.lucent.com/~quantum -(O)-==-o\ [EMAIL PROTECTED] - http://www.gaunt.org [EMAIL PROTECTED] wrote: > > In a message dated: Tue, 20 Aug 2002 16:43:36 EDT > [EMAIL PROTECTED] said: > > > I believe it was Ken Thompson, and I believe the remark was intended to be > >humorous. Step back and ask: Why would he spell "create" as "creat" in the > >first place? If you are going to type five characters, you might as well > >type six. The reason it was spelled "creat" in the first place was the > >linked only supported five characters. That has caused much > >head-scratching, question-asking, and recompiling-due-to-typos; hence the > >remark about the spelling. > > Okay, I'll buy that, but why create a linker that only supports 5 > character function names? What was it which caused this scenario? > I guess I don't fully understand the role of linkers, if this should > be an obvious thing :) > -- > > Seeya, > Paul > -- > It may look like I'm just sitting here doing nothing, >but I'm really actively waiting for all my problems to go away. > > If you're not having fun, you're not doing it right! > > ___ > gnhlug-discuss mailing list > [EMAIL PROTECTED] > http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: sorting pathnames by basename
The complexity of a computer language can result in the use of language subsets. Specifically, if the language has too many features, programmers will only learn a subset and since different programmers will learn different subsets, it becomes difficult for language implementors to guarantee that the language will be consistent across implementations. Niklaus Wirth, the designer of Pascal, operated under the philosophy that computer languages should be no more complicated than necessary. Later iterations of Pascal such as Modula 2 and Oberon (and their variations) were made succesively simpler in order to allow the programmer to accomplish his programming goals while at the same time allowing those that needed to read these programs to be able to clearly understand them (based on simplified syntax and a smaller number of reserved words). The Oberon language grammer fits nicely on a double sided 8x10 sheet of paper. The entire language report only occupies around 20 pages. See the language report at: http://www.oberon.ethz.ch/oreport.html and take note of the quote from A. Einstein ;^) -Alex P.S. Specific languages that have been described as too complex - PL/I, Ada, C++... - Original Message - From: "Erik Price" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Wednesday, August 21, 2002 1:32 AM Subject: Re: sorting pathnames by basename > > On Tuesday, August 20, 2002, at 08:28 PM, [EMAIL PROTECTED] wrote: > > > All three mean the same thing, but the first is by far the most > > common in > > American English. Perl is a lot like English. > > I couldn't agree more. Here's why: > > English is supposedly the hardest language in the world to learn. Why? > Because there are so many ways to say the same thing! Yes, that is the > reason AFAIK. I'm not just talking about synonyms, I'm talking about > the way you can structure the grammar of the language to mean the same > thing, but in so many ways. > > I studied Japanese for years when I was in college, almost > double-majoring with it but not quite. And I was struck by how > systematic that language is. Sure, a lot of people will naturally > wonder if it's very hard, since there is a complex writing system based > on Chinese that consists of tens of thousands of characters (the > Kanji). But as far as learning the core, the very grammar of the > language itself, it is amazingly straightforward! Although there is > "more than one way to do it", this generally has more to do with slang > and formality than with the way that the language is structured. > > Perl is definitely English-like in this regard. > > > > > > Erik > > > > > > -- > Erik Price > > email: [EMAIL PROTECTED] > jabber: [EMAIL PROTECTED] > > ___ > gnhlug-discuss mailing list > [EMAIL PROTECTED] > http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss > ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: Perl (or Unix vs. MS, actually)
In a message dated: Tue, 20 Aug 2002 16:52:35 EDT
"Steven W. Orr" said:
>On Tue, 20 Aug 2002 [EMAIL PROTECTED] wrote:
>=>In a message dated: Tue, 20 Aug 2002 15:16:58 CDT
>=>Thomas Charron said:
>=>
>=>For example, in shell, the construct:
>=>
>=> cd /tmp && rm foo
>
>Whotchoo talkin 'bout Willis?
>
>cd == chdir is a builtin command. But point taken.
Okay, bad example :) How about:
PID=`ps -ef | grep foo | awk '{print $2}'`
kill -9 $PID
That's 4 sub-processes :)
>Actually, debugging bash isn't all that bad.
I'm not saying it can't be done, but...
>set -x
isn't the same as a debugger with break points, watches, etc.
>And there is a full blown
>debugger available for ksh with breakpoints and everything. :-)
Ooh, now *that's* cool! Thanks!
--
Seeya,
Paul
--
It may look like I'm just sitting here doing nothing,
but I'm really actively waiting for all my problems to go away.
If you're not having fun, you're not doing it right!
___
gnhlug-discuss mailing list
[EMAIL PROTECTED]
http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: UNIX Arcana [was Re: Perl (or Unix vs. MS, actually) ]
In a message dated: Tue, 20 Aug 2002 16:43:36 EDT [EMAIL PROTECTED] said: > I believe it was Ken Thompson, and I believe the remark was intended to be >humorous. Step back and ask: Why would he spell "create" as "creat" in the >first place? If you are going to type five characters, you might as well >type six. The reason it was spelled "creat" in the first place was the >linked only supported five characters. That has caused much >head-scratching, question-asking, and recompiling-due-to-typos; hence the >remark about the spelling. Okay, I'll buy that, but why create a linker that only supports 5 character function names? What was it which caused this scenario? I guess I don't fully understand the role of linkers, if this should be an obvious thing :) -- Seeya, Paul -- It may look like I'm just sitting here doing nothing, but I'm really actively waiting for all my problems to go away. If you're not having fun, you're not doing it right! ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: sorting pathnames by basename
On 21 Aug 2002, at 8:52am, Kevin D. Clark wrote: >> ... sort -fdt/k1 ... > > ... Those options to sort don't work on my Linux box ... I assume you mean the -k switch? On my RHL 7.3 box, which has GNU sort from GNU textutils 2.0.21, the "info" manual for "sort" states: > Warning: the `+POS1' usage is obsolescent. Future versions of > POSIX will require that support for it be withdrawn. Use `--key' > (`-k') instead. Just FYI. -- Ben Scott <[EMAIL PROTECTED]> | The opinions expressed in this message are those of the author and do not | | necessarily represent the views or policy of any other person, entity or | | organization. All information is provided without warranty of any kind. | ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: sorting pathnames by basename
On Wed, Aug 21, 2002 at 08:52:14AM -0400, Kevin D. Clark <[EMAIL PROTECTED]> wrote: > Bob Bell <[EMAIL PROTECTED]> writes: > > sed -e 's;\(\(.*\)/\)*\(.*\);\3/\1\3;' | sort -fdt/k1 | sed -e 's;[^/]*/\(.*\);\1;' > > This doesn't work for me for a number of reasons: > > 1: sed is known to contain a number of bugs, and you're hitting one > of them on my SVR4-derived Solaris box. That first call to sed > doesn't yield any changes. Well, I can't be responsible for broken Unixen, now, can I? Fix your box :-) > 2: Those options to sort don't work on my Linux box. Really? They work on mine. What `sort` do you have? Mine's from a textutils-2.0e-8 RPM. All options I used are part of UNIX 98 (http://www.opengroup.org/onlinepubs/007908799/xcu/sort.html). > Still, I see where you're going here. How about this? > > sed -e 's;\(.*\)/\([^/]*\)$;\2/\1/\2;' | sort -t / +0 | sed 's;[^/]*/;;' > > This works on the various Unix boxen that I have at my disposal. Nope, you've just forced the pathname to contain a '/'. But here's a simplified version that I think will work around your issues for non-UNIX 98 compliant machines: sed -e 's;\(.*/\)*\(.*\);\2/\1\2;' | sort -f -d -t / +0 | sed -e 's;[^/]*/;;' If '-f' and '-d' are problems, you could just drop them. I used them because they were in the original suggestion, so I figure MOD wanted to use them in the comparison. -- Bob Bell <[EMAIL PROTECTED]> - "Beware of the above code; I have only proved it correct, not tried it." -- Donald Knuth, famous computer scientist ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: sorting pathnames by basename
Bob Bell <[EMAIL PROTECTED]> writes: > '/' is an illegal character in a filename, right? > > sed -e 's;\(\(.*\)/\)*\(.*\);\3/\1\3;' | sort -fdt/k1 | sed -e 's;[^/]*/\(.*\);\1;' This doesn't work for me for a number of reasons: 1: sed is known to contain a number of bugs, and you're hitting one of them on my SVR4-derived Solaris box. That first call to sed doesn't yield any changes. 2: Those options to sort don't work on my Linux box. Still, I see where you're going here. How about this? sed -e 's;\(.*\)/\([^/]*\)$;\2/\1/\2;' | sort -t / +0 | sed 's;[^/]*/;;' This works on the various Unix boxen that I have at my disposal. Regards, --kevin -- Kevin D. Clark / Cetacean Networks / Portsmouth, N.H. (USA) cetaceannetworks.com!kclark (GnuPG ID: B280F24E) alumni.unh.edu!kdc ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
